Welcome!
I've got large test database (>20000 entries), set up replication using syncrepl (refreshAndPersist). It work's fine when consumer is connected to provider while importing all database. But when I add new consumer to use existing database only first 500 records are replicated. I know about sizelimit, but I don't want to raise it (because I don't know how big my database will grow). The question is - why consumer won't get rest of the entries in another queries? What am I doing wrong? I can't import all database everytime I connect new consumer in a production enviroment.
On Fri, 17 Oct 2008, Łukasz Wąsikowski wrote:
about sizelimit, but I don't want to raise it (because I don't know how big my database will grow). The question is - why consumer won't get
Of course you don't want to specify an explicit number, which is why there's the keyword "none." Try:
limits dn.exact="cn=syncreplConsumer,dc=example,dc=com" size=none
rest of the entries in another queries? What am I doing wrong? I can't
I don't think the syncrepl client knows about paged results or anything like that. I've only seen it work with one big search.
--On Friday, October 17, 2008 10:20 AM +0200 Łukasz Wąsikowski lukasz@wasikowski.net wrote:
Welcome!
I've got large test database (>20000 entries), set up replication using syncrepl (refreshAndPersist). It work's fine when consumer is connected to provider while importing all database. But when I add new consumer to use existing database only first 500 records are replicated. I know about sizelimit, but I don't want to raise it (because I don't know how big my database will grow). The question is - why consumer won't get rest of the entries in another queries? What am I doing wrong? I can't import all database everytime I connect new consumer in a production enviroment.
slapcat the master, then slapadd the brand new consumer.
I highly advise you read the admin guide section on replication. Aaron already provided you with the answer to your question about sizelimits (use the limits command to adjust for the consumer's search ID).
http://www.openldap.org/doc/admin24/replication.html#Pull%20Based http://www.openldap.org/doc/admin24/replication.html#Configuring%20the%20different%20replication%20types
--Quanah
--
Quanah Gibson-Mount Principal Software Engineer Zimbra, Inc -------------------- Zimbra :: the leader in open source messaging and collaboration
On Friday 17 October 2008 18:19:41 Quanah Gibson-Mount wrote:
--On Friday, October 17, 2008 10:20 AM +0200 Łukasz Wąsikowski
lukasz@wasikowski.net wrote:
Welcome!
I've got large test database (>20000 entries), set up replication using syncrepl (refreshAndPersist). It work's fine when consumer is connected to provider while importing all database. But when I add new consumer to use existing database only first 500 records are replicated. I know about sizelimit, but I don't want to raise it (because I don't know how big my database will grow). The question is - why consumer won't get rest of the entries in another queries? What am I doing wrong? I can't import all database everytime I connect new consumer in a production enviroment.
slapcat the master, then slapadd the brand new consumer.
This would be faster. But, it is entirely plausible to want to initialise a consumer without access to the database files on the provider.
I highly advise you read the admin guide section on replication. Aaron already provided you with the answer to your question about sizelimits (use the limits command to adjust for the consumer's search ID).
Using slapcat|slapadd to initialise the directory can in fact hide problems with size limits ...
Regards, Buchan
Hi!
Aaron Richton pisze:
about sizelimit, but I don't want to raise it (because I don't know how big my database will grow). The question is - why consumer won't get
Of course you don't want to specify an explicit number, which is why there's the keyword "none." Try:
limits dn.exact="cn=syncreplConsumer,dc=example,dc=com" size=none
Thank you, that was it.
Łukasz Wąsikowski wrote:
Hi!
Quanah Gibson-Mount pisze:
slapcat the master, then slapadd the brand new consumer.
AFAIR slapcat shouldn't be used on running slapd, right?
No. You *can safely* run slapcat while slapd is running. You *can't* run slapadd and slapindex while slapd is running.
p.
Ing. Pierangelo Masarati OpenLDAP Core Team
SysNet s.r.l. via Dossi, 8 - 27100 Pavia - ITALIA http://www.sys-net.it ----------------------------------- Office: +39 02 23998309 Mobile: +39 333 4963172 Fax: +39 0382 476497 Email: ando@sys-net.it -----------------------------------
openldap-software@openldap.org
-
Aaron Richton -
Buchan Milne -
Pierangelo Masarati -
Quanah Gibson-Mount -
Łukasz Wąsikowski